yak(coverage-floors-re-measure-and-re-record-after-the-behavior-test-yaks-mgr8): coverage floors: re-measure and re-record after the behavior-test yaks

app/mix.exs

@@ -192,12 +192,14 @@ defmodule Firehose.MixProject do
   # land; never lower it. A drop means the change (or an exclusion) is wrong,
   # not the floor.
   #
-  # Measured 2026-10-28 (PostgreSQL on localhost:5432, exclusions below):
-  # 88.19% after the web-controller/auth edge-case behavior tests (87.56% after
-  # the editorial-LiveView ones, 86.97% before them, and 86.79% / 86.17% at
-  # earlier exclusion revisions), identical via `mix test --cover` and via
+  # Measured 2026-10-29 (PostgreSQL on localhost:5432, exclusions below):
+  # 94.01% on `main` = 9c0f33e, with boundary enforcement, credo, the mutation
+  # scope and the core-components behavior tests all landed (88.19% after the
+  # web-controller/auth edge-case tests, 87.56% after the editorial-LiveView
+  # ones, 86.97% before them, and 86.79% / 86.17% at earlier exclusion
+  # revisions), identical via `mix test --cover` and via
   # `mix test --cover --export-coverage default` + `mix test.coverage`. Floor is
-  # the integer below that — so a floor of 88 means "at least 88.00%".
+  # the integer below that — so a floor of 94 means "at least 94.00%".
   #
   # `test.coverage` deletes stale exports before sampling: `mix test.coverage`
   # imports *every* .coverdata it finds, so leftover exports from earlier runs
@@ -243,7 +245,7 @@ defmodule Firehose.MixProject do
   # that is where the behavior-test work shows up as a higher floor.
   defp test_coverage do
     [
-      summary: [threshold: 88],
+      summary: [threshold: 94],
       ignore_modules: [
         Firehose.Repo,
         Firehose.Mailer,

blogex/mix.exs

@@ -62,22 +62,22 @@ defmodule Blogex.MixProject do
   # in blogex/). Rule: the threshold is a FLOOR. Raise it as behavior-test yaks
   # land; never lower it. A drop means the change (or an exclusion) is wrong.
   #
-  # Measured 2026-10-28, with the test scaffolding below excluded: 99.69% — the
-  # same via `mix test --cover` and via `mix test --cover --export-coverage
+  # Measured 2026-10-29, with the test scaffolding below excluded: 100.00% —
+  # the same via `mix test --cover` and via `mix test --cover --export-coverage
   # default` + `mix test.coverage`. Floor is the integer below that.
   #
   # Excluded: the test scaffolding in `ignore_modules` below. The two registry
   # blog doubles (Blogex.Test.AlphaBlog/BetaBlog) are deliberately not listed
-  # there: they are fully exercised, and excluding them measures 99.69% too —
-  # the 99 floor holds either way.
+  # there: they measure 100.00% too, so counting them costs nothing.
   #
-  # Everything in lib/blogex counts, and one gap is left:
-  # Blogex.LinkError is at 88.89% (8 of 9 lines). The rendered-HTML behavior
-  # tests closed the rest: Blogex.Components and Blogex.Layout went from 0% to
-  # 100%, and Blogex.Router now covers its browser arms and `match _`.
+  # Everything in lib/blogex counts, and nothing is uncovered: Blogex.LinkError
+  # was the last gap at 88.89% (8 of 9 lines) until the credo yak's tests closed
+  # it. The rendered-HTML behavior tests took Blogex.Components and Blogex.Layout
+  # from 0% to 100%, and Blogex.Router covers its browser arms and `match _`.
+  # A floor of 100 means every new line needs a test before `make verify` passes.
   defp test_coverage do
     [
-      summary: [threshold: 99],
+      summary: [threshold: 100],
       ignore_modules: [
         Blogex.Test.FakeBlog,
         Blogex.Test.PostBuilder,